摘要： 基于内容的语义Web检索只考虑内容本身，没有考虑用户的不同，不能准确反映用户需求。为此，提出一个自适应语义Web检索框架，对于Web中文文档，借助HowNet知识库给出一种本体学习方法，通过提取用户客观、显式和隐式信息建立用户信息库，并设计用户初始查询本体和个性化查询本体构建算法，从而实现用户的自适应检索。实验结果表明，该方法具有较高的检索效率。

关键词: 自适应, 本体学习, 个性化本体, 语义Web检索, 用户信息库

Abstract: Content-based Semantic Web retrieval only considers the content itself, without taking into account the different users, and it can not accurately reflect user needs. Aiming at the problem, this paper proposes a framework for adaptive semantic Web retrieval. For Chinese documents on the Web, it gives a kind of ontology learning method with HowNet, builds user information database by extracting the user objective, explicit and implicit information, designs the initial query ontology and personalized query ontology construction algorithm, and ultimately achieves the user’s adaptive retrieval. Result shows the method has higher retrieval efficiency.

Key words: adaptive, ontology learning, personalized ontology, semantic Web retrieval, user information base

中图分类号: 

• TP391